Condition monitoring of wind turbine based on incremental learning and multivariate state estimation technique

Ziqi Wang, Changliang Liu and Feng Yan

Renewable Energy, 2022, vol. 184, issue C, 343-360

Abstract: With the development of wind turbine (WT) operation and maintenance technologies, the condition monitoring (CM) method based on the data of supervisory control and data acquisition (SCADA) systems has become one of the most popular and cheapest ways to detect the anomalies. As the operation condition of WT will change over time, it is necessary to real-time update the CM model with little manual maintenance to ensure its long-term performance. Therefore, a SCADA data-driven WTCM method based on incremental learning and multivariate state estimation technique (MSET) is proposed. Firstly, for the memory matrix (MM) of MSET, a similarity-based sample selection method is proposed, which is simpler and more effective than the previous method. Secondly, based on the proposed sample selection method, an incremental learning strategy for MSET is proposed, which can add the normal new data in MM and remove the redundant data in real-time. Aiming at the slow computation speed caused by the large MM, a dynamic down-sampling method is proposed, which makes only half of the data participate in the real-time computation and can reduce the computation time by at least 80%. About three years SCADA data of two different types of WTs are used to verify the proposed method. The experimental results show that the incremental MSET can maintain higher estimation accuracy and lower false alarm rate in long-term operation, and can detect the potential gearbox faults hours to weeks in advance.

Keywords: Wind turbine; Condition monitoring; Incremental learning; Multivariate state estimation technique (MSET); Supervisory control and data acquisition (SCADA) (search for similar items in EconPapers)
Date: 2022
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (2)

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0960148121016542
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:renene:v:184:y:2022:i:c:p:343-360

DOI: 10.1016/j.renene.2021.11.071

Access Statistics for this article

Renewable Energy is currently edited by Soteris A. Kalogirou and Paul Christodoulides

More articles in Renewable Energy from Elsevier
Bibliographic data for series maintained by Catherine Liu ().